Ensuring academic integrity and assessment security with
redesigned online delivery
The following guide is designed to assist unit chairs with
redesigning assessment, to suit fully online delivery, without
invigilated exams.
Some assessments can easily move to online delivery or
submission some assessment will need to be redesigned to ensure
that quality and rigour are maintained.
Contents
1. Academic integrity, assessment security and digital
assessment – outlines the latest research on how to keep digital
assessment tasks secure.
2. Three key questions to guide redesigning exams for online
delivery – provides prompts to help unit chairs work through design
decisions
3. Redesigning exams: decision helper – provides a decision tree
to assist unit chairs move exams to new formats
4. Examples of converting single answer correct/MCQ questions to
a unique answer format
5. Tips for moving practical exams or assessments to online
equivalents
6. Tips for moving complex unique response exams (eg essay
style) to online equivalents
The research
The CRADLE team have conducted a range of projects on assessment
security and academic integrity, including work on detecting
contract cheating (including a CRADLE Suggests resource [link]),
the quality of contract cheated work, and the security of online
examinations.
The following advice comes from CRADLE research as well as other
cited sources from the literature.
1. Academic integrity, assessment security and digital
assessment
What is academic integrity and assessment security?
With a rapid shift to online learning, many educators have
raised concerns about student cheating. To address this, we need to
address academic integrity and assessment security. Academic
integrity focuses on equipping learners with the capabilities and
values necessary to conduct ethical scholarship. In contrast,
assessment security focuses on hardening assessment against
attempts to cheat, and on detecting any cheating that has occurred.
These different missions are both necessary to ensure that students
who complete university degrees have met the required outcomes.
Don’t assume that digital assessment is less secure
No assessment is immune to cheating. While in-person
examinations are often thought of as more secure, recent
large-scale survey research suggests that exams are the site of
both more cheating, and more undetected cheating, than take-home
written tasks (Harper, Bretag, & Rundle, 2020). The types of
assignments students say they are least likely to cheat on are
reflections on practicums, vivas, personalised and unique tasks,
and in-class tasks (Bretag et al., 2019). Some of these are
translatable into digital modes.
Focus on securing high-stakes tasks that matter
programmatically
Securing every act of assessment is infeasible, and would likely
lead to poor learning experiences for students. When choosing which
tasks to focus on, those that contribute to degree outcomes matter
most. If a particular outcome is assessed at first, second and
third year, it is probably much more important to secure the
third-year task. More resource-intensive approaches like vivas
might be feasible if applied sparingly to these programmatically
high-stakes moments of assessment. While cheating should not be
ignored in any task, lower-stakes tasks may provide opportunities
to instead focus on building cultures of integrity and trust.
Reconsider the need to assess low-level outcomes and tasks with
one right answer
Assessing recall of facts requires students to not have access
to those facts. Tasks with ‘one right answer’ rely on restricting
access to that answer, this is not possible in unsupervised digital
quizzes. While there are some circumstances where these types of
assessment are essential, if it is possible to substitute them with
tasks involving higher-level outcomes these may be vulnerable to
fewer types of cheating.
Vivas might improve assessment security
CRADLE recently conducted a study where we paid students to
cheat in a viva. We have not tried to publish this study yet
because the results look too good to be true: the Deakin sessional
markers we employed were able to spot cheating, every time. We
present this result with caution because we wish to replicate it
before we publish it. But there may be some benefit to vivas or
similar conversations with students about their work as a way to
improve assessment security. This includes formal vivas,
conversations over Zoom about an assignment a student has
submitted, or teacher-student interactions that are integrated
throughout the task.
Talk with teaching teams about cheating and integrity in digital
assessment
Panicking about cheating in digital assessment is unproductive
and not based in evidence. However, it is worthwhile having a think
and a chat about the different ways that integrity can be promoted
online, and the types of cheating to look out for. CRADLE research
has repeatedly found that being alerted to the possibility of
cheating is one of the most effective interventions available at
increasing detection rates (e.g. Dawson & Sutherland-Smith,
2018, 2019). As sessional staff often do much of the marking it is
important to involve them in these discussions.
Talk with students about the dangers of cheating
In addition to the usual academic integrity conversations you
might have, students also need to know about the risks to
themselves of cheating. Assignments bought online are often poor
quality, with one CRADLE study finding most purchased assignments
were not even of pass quality (Sutherland-Smith & Dullaghan,
2019). Universities take cheating very seriously, and penalties can
include exclusion. Even worse, students are sometimes blackmailed
by cheating services (Yorke, Sefcik, & Veeran-Colton, 2020). In
addition to educating students about the benefits of academic
integrity, students also need to know about the dangers of
cheating.
References
Bretag, T., Harper, R., Burton, M., Ellis, C., Newton, P., van
Haeringen, K., . . . Rozenberg, P. (2019). Contract cheating and
assessment design: exploring the relationship. Assessment &
Evaluation in Higher Education, 44(5), 676-691.
doi:10.1080/02602938.2018.1527892
Dawson, P. (2016). Five ways to hack and cheat with
bring-your-own-device electronic examinations. British Journal of
Educational Technology, 47(4), 592-600. doi:10.1111/bjet.12246
Dawson, P., & Sutherland-Smith, W. (2018). Can markers
detect contract cheating? Results from a pilot study. Assessment
& Evaluation in Higher Education, 43(2), 286-293.
doi:10.1080/02602938.2017.1336746
Dawson, P., & Sutherland-Smith, W. (2019). Can training
improve marker accuracy at detecting contract cheating? A
multi-disciplinary pre-post study. Assessment & Evaluation in
Higher Education, 44(5), 715-725.
doi:10.1080/02602938.2018.1531109
Harper, R., Bretag, T., & Rundle, K. (2020). Detecting
contract cheating: examining the role of assessment type. Higher
Education Research & Development, 1-16.
doi:10.1080/07294360.2020.1724899
Sutherland-Smith, W., & Dullaghan, K. (2019). You don’t
always get what you pay for: User experiences of engaging with
contract cheating sites. Assessment & Evaluation in Higher
Education, 44(8), 1148-1162. doi:10.1080/02602938.2019.1576028
Yorke, J., Sefcik, L., & Veeran-Colton, T. (2020). Contract
cheating and blackmail: a risky business? Studies in Higher
Education, 1-14. doi:10.1080/03075079.2020.1730313
2. Three key questions to guide redesigning exams for online
delivery
The following three prompts may help explain some of the
decisions facing you in moving from an invigilated exam to a new
digital assessment task. Refer also to the Redesigning exams
decision helper resource. They focus on how small changes can
maximise quality and assessment security, so that new tasks can be
run in online formats without invigilation.
Is this exam particularly critical?
Essential exams primarily take place in final years, and assure
that the student has met some critical course learning outcomes. If
you think your exam is absolutely critical for this purpose whereby
the student will not be able to practice as a professional without
doing a version of this exam before graduation, then we suggest you
contact your ADTL to make a case for an invigilated online
examination. This is for exceptional circumstances only.
Does the exam focus on knowledge recall?
Exams that primarily test knowledge recall or have a single
correct answer, such as multiple-choice questions, make cheating
easy where invigilation is not possible (see CRADLE Suggests -
Academic integrity, assessment security and digital
assessment).
We recommend that you find alternatives for exams that test
knowledge recall and/or exams with single correct answers such as
multiple choice.
One alternative is to provide scheduled online tasks with timed
release – but with redesigned questions which test students on
their reasoning behind an answer choice. An added benefit is that
these formats engage students in deeper learning. Keep in mind that
students may need more time to complete these more difficult
assessments and marking will take longer. See Redesigning exams
decision helper for extra assistance.
Other exams will easily translate to the online environment.
These include take-home or open-book/open-web exams, vivas or exam
formats that involve unique answers, creativity and
problem-solving. We suggest these can run as take-home exams.
How will you communicate changes with students and help them
prepare?
It is important to let students know what the new tasks will
entail, why you are using the new format and to provide criteria
outlining expectations for achievement. Where this new format asks
them to do something they may not be used to – such as a
time-limited oral presentation – opportunities for practice and
additional resources will be required.
3. Redesigning exams: decision helper (for a screen readable
version please click here)
4. Examples of converting single answer correct/MCQ questions to
a unique answer format
Converting a single answer correct or MCQ question to a unique
answer format often extends what is being tested. In these
examples, you will see that not only do the students have to recall
or identify the correct answer but explain it, thereby testing
their ability to analyse/synthesise knowledge and articulate this
analysis to students.
These two examples illustrate an easy means to convert these
types of questions to a more unique answer format. Note that
students will take longer to do these and that they will take
longer to grade.
Example 1: Knowledge recall multiple choice or single correct
answer
Original multiple choice question
The strongest and most resilient connective tissue is
A. adipose tissue
B. reticular connective tissue
C. fibrocartilage tissue
D. elastic connective tissue
E. areolar connective tissue
Original single correct question
What is the strongest and most resilient connective tissue?
Answer: fibrocartilage tissue
Alternative questions, focussing on explanation:
Why is fibrocartilage tissue the strongest and most resilient
connective tissue?
Comparing adipose tissue and fibrocartilage tissue, discuss
reasons for relative strength and resilience of these connective
tissues.
Example 2: Analytic style multiple choice question or short
answer
Original multiple choice question
In a study aimed at identifying factors associated with risk of
developing dementia, a group of elderly people with a formal
diagnosis of dementia were compared with a group of elderly people
without dementia for a range of factors related to health,
lifestyle and occupation. The patients with dementia were matched
with those without dementia by age, sex and area of residence. Data
collection was by interview. For the patients with severe dementia,
where the dementia interfered with data collection, surrogates
(usually a family member) assisted with data collection.
This study is a
a)Case-control study
b)Cohort study
c)Cross-sectional survey
d)Field study
Original single correct question
In a study aimed at identifying factors associated with risk of
developing dementia, a group of elderly people with a formal
diagnosis of dementia were compared with a group of elderly people
without dementia for a range of factors related to health,
lifestyle and occupation. The patients with dementia were matched
with those without dementia by age, sex and area of residence. Data
collection was by interview. For the patients with severe dementia,
where the dementia interfered with data collection, surrogates
(usually a family member) assisted with data collection. What type
of study is this?
Answer: case-control study
Alternative questions, focussing on explanation:
Harder
In a study aimed at identifying factors associated with risk of
developing dementia, a group of elderly people with a formal
diagnosis of dementia were compared with a group of elderly people
without dementia for a range of factors related to health,
lifestyle and occupation. The patients with dementia were matched
with those without dementia by age, sex and area of residence. Data
collection was by interview. For the patients with severe dementia,
where the dementia interfered with data collection, surrogates
(usually a family member) assisted with data collection.
What type of study is this? Why do you think this?
Easier
In a case-control study aimed at identifying factors associated
with risk of developing dementia, a group of elderly people with a
formal diagnosis of dementia were compared with a group of elderly
people without dementia for a range of factors related to health,
lifestyle and occupation. The patients with dementia were matched
with those without dementia by age, sex and area of residence. Data
collection was by interview. For the patients with severe dementia,
where the dementia interfered with data collection, surrogates
(usually a family member) assisted with data collection.
What makes this a case control study?
Acknowledgement for original multiple choice questions: Jennifer
Lindley, Monash University.
5. Tips for moving practical exams or assessments to online
equivalents
Types of practical assessments include:
· Laboratory-based practicals (eg chemistry, physics, health
sciences etc).
· Performance-based assessments (eg fine arts, dance, etc).
· Physical artefact development (eg engineering, fine arts
etc).
· Psychomotor skills (eg physiotherapy, nursing and other health
professions etc).
· Interpersonal skills (eg medicine and other health
professional consultation skills etc).
· Language skills (eg vivas).
Suggested alternatives:
· Video-based uploads of tasks performed in home environment
using CloudDeakin.
· Online simulation-based tasks (eg pre-existing computer-based
sims) – might be worth coordinating with other universities to
share simulation resources.
· Providing a portfolio rather than making a single piece of
work in a scheduled time frame (eg a series of videos showing
development of an artwork/artefact).
· Critique and explanation of video practice – need to find or
make videos and post online plus design a critique task for
students.
· Real time observed practicals/vivas (very resource intensive)
– zoom or BBcollaborate.
In some cases, it may be necessary to defer assessment until a
later point in the course.
Please refer to Faculty teams who can create videos and provide
advice on file formats and support documentation for both students
and staff
6. Tips for moving complex unique response exams (eg essay
style) to online equivalents
Types of exams include:
· Essays (eg history, English, visual arts, most arts based
subjects etc).
· Problems with individualised solutions (eg computer science
programming, some mathematical problems, engineering etc).
· Paper-based authentic artefact development (eg marketing
plans, legal response).
· Take-home (open book) exams (eg law).
These exams are relatively easy to translate to online
environments. Some suggestions:
· Take home exams can be easily transferred online – these do
not need to change.
· Longer form essays and problems designed for an exam situation
can be translated to a take-home, open-book, exam but these will
need to be modified to accommodate a shift from focused analysis
and application to demonstration of a research-based approach.
Adding a process or reflective component can help this
adjustment.
· Other tasks may need to be rewritten so that they can still
work to ensure the student has access to all resources.
· If keeping as scheduled on-line task with timed release. Add
reflective components to help avoid cheating: eg ask students to
upload videos explaining their answers and grade these; ask
students to reflect on an online discussion that has been had
exclusively in class.
· If able to resource an online viva format using zoom or
collaborate may work, where students are asked questions and
respond on the spot – feedback and grades can be done in real time
to reduce time in grading.
10
Circulation for comment draft March 2020. This resource has
produced as a collaboration between CRADLE and Deakin Learning
Futures (DLF).